FEBS Letters 588 (2014) 3726–3731

journal homepage: www.FEBSLetters .org

Heterologous expression of AtPAP2 in transgenic potato influencescarbon metabolism and tuber development

http://dx.doi.org/10.1016/j.febslet.2014.08.0190014-5793/� 2014 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

⇑ Corresponding authors. Fax: +49 (0)331 567 8250 (A.R. Fernie). Fax: +85225599114 (B.L. Lim).

E-mail addresses: Fernie@mpimp-golm.mpg.de (A.R. Fernie), bllim@hku.hk(B.L. Lim).

Youjun Zhang a,b, Feng Sun b, Joerg Fettke c, Mark Aurel Schöttler a, Lawrence Ramsden b,Alisdair R. Fernie a,⇑, Boon Leong Lim b,⇑a Max Planck Institute of Molecular Plant Physiology, Am Mühlenberg 1, 14476 Potsdam-Golm, Germanyb School of Biological Sciences, The University of Hong Kong, Pokfulam, Hong Kong, Chinac Institute of Biochemistry and Biology, University of Potsdam, Golm, Germany

a r t i c l e i n f o

Article history:Received 16 June 2014Revised 14 August 2014Accepted 16 August 2014Available online 27 August 2014

Edited by Ulf-Ingo Flügge

Keywords:PotatoAtPAP2PhotosynthesisTuber yieldSugar efflux

a b s t r a c t

Changes in carbon flow and sink/source activities can affect floral, architectural, and reproductivetraits of plants. In potato, overexpression (OE) of the purple acid phosphatase 2 of Arabidopsis(AtPAP2) resulted in earlier flowering, faster growth rate, increased tubers and tuber starch content,and higher photosynthesis rate. There was a significant change in sucrose, glucose and fructoselevels in leaves, phloem and sink biomass of the OE lines, consistent with an increased expressionof sucrose transporter 1 (StSUT1). Furthermore, the expression levels and enzyme activity ofsucrose-phosphate synthase (SPS) were also significantly increased in the OE lines. These findingsstrongly suggest that higher carbon supply from the source and improved sink strength can improvepotato tuber yield.� 2014 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction

Photosynthate supply and sink strength have been demon-strated experimentally to be the major determinants of crop yield[1]. Assimilated carbon from photosynthesis supplies both energysources for metabolism and building blocks for complex carbohy-drates. Photoassimilate is further partitioned within the mesophyllcells and transported, mainly in the form of sucrose, from source tosink tissues to support plant growth and development. The plantgrowth rate depends on the photosynthetic fixation capacity andon how efficiently the fixed carbon is utilized in biosyntheticprocesses which support growth. Accordingly, the distribution ofcarbon assimilates assists in balancing photosynthetic activity inthe source leaves and photoassimilate utilization and storage insinks. During this process SPS activity, which usually determinessucrose synthesis rates, and sucrose transporters (SUTs) requiredfor sucrose phloem loading, have been demonstrated to beregulated by protein phosphorylation status [2].

Purple acid phosphatases (PAPs) are a family of acidic binuclearmetalloenzymes which hydrolyze phosphate esters and anhy-drides under acidic conditions. Many plant PAPs were shown tobe induced by Pi starvation and involved in phosphorus metabo-lism [3]. Our previous studies demonstrated that overexpressionof AtPAP2, an Arabidopsis PAP with an additional C-terminalhydrophobic motif, located at the outer membrane of both chloro-plasts and mitochondria [4], drastically enhanced the growth rateand seed yield of Arabidopsis thaliana and Camelina sativa [5,6].These results imply that AtPAP2 can potentially regulate plantcarbon metabolism.

2. Materials and methods

2.1. Plant materials and growth conditions

Potatoes (Solanum tuberosum var. Bintje) were provided by Prof.M. L. Chye of the University of Hong Kong and potatoes (S. tubero-sum cv. Desirée) were from the Max Planck Institute of MolecularPlant Physiology, Potsdam-Golm. Potato was maintained in tissueculture with 16-/8-h day/night cycles on Murashige and Skoogmedium [7] in growth room, which contained 2% (w/v) sucrose.The top 5 internode of one-month-old WT potatoes were usedfor transformation. In addition, transgenic potato plants were first

Y. Zhang et al. / FEBS Letters 588 (2014) 3726–3731 3727

grown in a growth chamber (150 lmol m�2 s�1 and 75% relativehumidity (RH)) under 16 h light (22 �C)/8 h dark (18 �C) light per-iod for several weeks before they were transferred to greenhouse(Light intensity varied between a minimum of 90 and a maximumof 200 lmol m�2 s�1 PPFD and 75% RH) with the day/nightcondition of Hong Kong and Germany and watered twice everyweek before tuber collection. Tubers of all plants were collectedafter growing in soil for about 4 months.

2.2. Extraction of total plant RNA and quantitative RT-PCR analysis

Total RNA was isolated from fresh leaves by using TRIzolreagent (Invitrogen). To generate full-length cDNA for quantitativeRT-PCR, reverse transcription was performed using the M-MLVreverse transcriptase (Promega, Hong Kong). Full length AtPAP2cDNA was amplified by Pfx DNA polymerase (Invitrogen) andsubcloned into pBA002 by XhoI and SacI (Table S1) for potatotransformation.

Quantitative RT-PCR reaction was carried out in the presence ofSYBR Green with HotGoldStar DNA polymerase (Eurogentec) inRotor Gene 3000 cycler (LTF Labortechnik) using Rotor Gene soft-ware (version 4.6.94). An aliquot of 0.2 ll cDNA of the 10 ll RTreaction was used for each reaction. Relative quantification oftranscript amounts was calculated in relation to the respectiveubiquitin transcript level and given as percentage of ubiquitin.Primers (Table S1) were designed according to published papers[8] and produce 50- to 150-bp amplicon using Primer5 software.Quantitative RT-PCR data were corrected by calculation of thePCR efficiency individually using the LinReg PCR software [9].

2.3. Agrobacterium-mediated transformation and Southern blottinganalysis

The full-length coding region of the AtPAP2 cDNA (AT1G13900)was subcloned into the binary vector pBA002 downstream of thecauliflower mosaic virus (CaMV) 35S promoter (pBA002-CaMV35:AtPAP2). The vector was then introduced into Agrobacterium tum-efaciens strain GV3101 and internodal explants from 4-week-oldWT plantlets were used for transformation [10]. Southern blotanalysis was carried out as described [11]. The probes wereSB-PAP2-f and SB-PAP2-r (Table S1).

2.4. Western blotting analysis

Potato leaves were finely ground in a 1.5-ml Eppendorf tubecontaining 200 ll of ice-cooled extraction buffer (50 mM Tris–HCl, pH7.4 containing 150 mM NaCl, 1 mM EDTA, 0.2 mM PMSF)and incubated on ice for 30 min with occasional mixing. The pro-tein extract was separated by centrifugation at 10000�g for30 min at 4 �C. The supernatant was transferred to a new 1.5-mlEppendorf tube and the protein concentration was determinedby the Bradford assay method using the Bio-Rad Protein AssayKit. Proteins (25 lg) were resolved by SDS–PAGE, transferred toHybond-C nitrocellulose membranes, immunodetected and thenthe proteins were visualized by the Enhanced Chemiluminescence(ECL) method (Amersham Biosciences).

2.5. Measurement of leaf assimilation rate

The leaf assimilation rates of potato were measured using a por-table photosynthesis system (LI-COR, LI-6400, Nebraska, USA) inthe morning (8.30 to 12:30 AM) under a fixed blue-red light-emit-ting diode (LED) light source. Nine measurements were made foreach three fully expanded intact leaves from the tip of 65 to67-day-old potato and at least 3 plants of each line were used formeasurement. Light curves were measured on 6 cm2 leaf area

using the instrument’s auto program function. Measurements weretaken in darkness, to determine leaf respiration, and at actinic lightintensities of 0, 125, 250, 500, 750 and 1000 lmol m�2 s�1 at 25 �Ccuvette temperature and a CO2 concentration of 400 ppm. Relativehumidity was set to 75%.

2.6. HPLC analysis of sugar content

For measurement of sucrose, glucose and fructose of plant tis-sues, an aliquot of 0.1 g freeze-dried tissue powder was dissolvedin 1 ml of 70% (v/v) ethanol, incubated at 70 �C for 90 min and cen-trifuged at 13000�g for 10 min. After passing through a 0.22 mm fil-ter, A volume of 10 ll sample was injected into a CarboPac PA 1column (4 � 250 mm) connected to a Dionex LC 20 Chromatographysystem and the sugar contents were analyzed by high performanceanion exchange chromatography with pulsed amperometric detec-tion (HPAEC-PAD) [12]. Standard curves were prepared by 0.0–0.1 mg/ml sucrose, fructose and glucose in 70% ethanol. The levelsof starch in the tubers were determined as described previously [13].

2.7. SPS activity assay

SPS activity was assayed by the anthrone test [14]. Sampleswere incubated for 20 min at 25 �C in 50 ll pre-balanced buffer(50 mM HEPES-KOH pH 7.5, 20 mM KCl, and 4 mM MgCl2) contain-ing (a) Vmax assay: 12 mM UDP-Glc and 10 mM Fru6P (in a 1:4 ratiowith glucose-6-phosphate (Glc6P), (b) Vlimiting assay: 4 mM UDP-Glc and 2 mM Fru6P (in a 1:4 ratio with Glc6P) and 5 mM KH2PO4.

2.8. Pull-down assay

Plant materials were ground in liquid nitrogen and incubated inice-cold buffer (50 mM HEPES-KOH pH 7.5, 150 mM NaCl, 50 mMNaF, 5 mM NaPPi, 1 mM dithiothreitol, 1 mM PMSF, 1 mM ben-zamidine) for 15 min and then they were centrifuged at12000�g for 20 min. One milligram extracted protein was addedto 5 lg GST-14-3-3 [15] on 50 ll GST beads with gentle agitationfor 2 h at 4 �C. The beads were then centrifuged at 2000�g for2 min at 4 �C and washed five times with pull-down washing buf-fer (50 mM HEPES-KOH pH 7.5, 150 mM NaCl, 1 mM dithiothrei-tol). The washed beads were eluted (50 mM Tris–HCl, 10 mMreduced glutathione, pH 8.0, 150 mM NaCl) and the supernatantwas subjected to SDS–PAGE for silver-staining and Westernblotting.

2.9. Co-immunoprecipitation (Co-IP) assay

Twenty microliter of soluble proteins in inhibitor buffer (1�PBS, pH 7.4, 1 mM PMSF, 1 mM NaVO4, 50 mM NaF, with COM-PLETE™ Protease Inhibitor Cocktail (Roche)) was incubated with10 lg first antibody for 2 h at 4 �C, and then mixed with 20 ll Pro-tein-G Sepharose (10 ml of 50% slurry in 1� PBS/20% ethanol (Gen-Script Co.) for 2 h. The pellets were washed three times in 1 mlinhibitor buffer and boiled in 20 ll SDS sample buffer [16] andused for SDS–PAGE and Western blotting.

3. Results and discussion

Overexpression of AtPAP2 in Arabidopsis and Camelina hasbeen shown to improve plant growth and seed yield [5,6]. Potatoesfeature a very large sink organ (tuber) for carbon and utilize a well-defined apoplastic phloem loading mechanism with sucrose trans-porters, and are highly suitable for further investigations of thebiological functions of AtPAP2. We generated four transgenic linesin potatoes with high and mild expressions of AtPAP2 proteins(Figs. S1 and S2). All three high AtPAP2 expressing potato lines

ig. 1. Growth phenotypes of 40-day-old plants. All three high AtPAP2 OE linesroduced more branches than the WT.

3728 Y. Zhang et al. / FEBS Letters 588 (2014) 3726–3731

Fp

(OE1, OE4 and OE7) grew faster and produced more lateralbranches than the WT and the mild AtPAP2 overexpression line(OE10; Figs. 1 and S3). The first harvest plants were grown in

Table 1Flowering time of the transgenic potatoes.

WT Vector O

S. tuberosum var. BintjeFlowering time (day) 81 ± 0.7a 89 ± 0.5b 4

WT OE1 O

S. tuberosum cv. DesiréeFlowering time (day) 54.5 ± 1.6a 38 ± 1.6b 3

Second generation potatoes (S. tuberosum var. Bintje) were grown from September 2010 twere grown under LD conditions [16 h light period (100–200 lE m�2 s�1), 25 �C and 8 h dPhysiology]. Throughout the light–dark cycle relative humidity was kept at 50%. The va

Table 2Tuber yield of the transgenic potatoes.

WT Vector

Second generation (S. tuberosum var. Bintje, SD)Tuber number/plant 2a 2a

Average weight/tuber, FW 7.4 ± 1a 12.5 ± 1.3ab

Tuber weight/plant, FW 14.8 ± 2a 25.1 ± 2.6ab

Root biomass/plant, DW 0.4 ± 0.1a 0.3 ± 0.1a

Aboveground biomass /plant, DW 2.9 ± 0.3a 2.9 ± 0.7a

Leaf biomass/Plant, DW 1.8 ± 0.3a 1.8 ± 0.6a

Petioles biomass/plant, DW 1.1 ± 0.2a 1.1 ± 0.2a

WT OE1

Third generation (S. tuberosum var. Bintje, LD)Tuber number/plant 3 15Tuber weight (g)/plant, FW 107.5 258.5

WT

Tuber yield and starch content of a different cultivar (S. tuberosum cv. Desirée, LD)Tuber number/plant 6.3 ± 1.4a

Tuber weight (g/plant), FW 176.0 ± 20.7a

Tuber weight (g/plant), DW 25.0 ± 2.9a

Tuber water content (mg/g), FW 857.8 ± 8.2a

Starch content (lmol Glucose/g), FW 351.7 ± 7.7a

Tubers > 1 g was measured. Samples were collected from 4 to 6 independent plants afterpotato (S. tuberosum var. Bintje) was grown in short day condition in Hong Kong and thePotatoes (S. tuberosum cv. Desirée) were grown under controlled conditions from Novemdarkness, 20 �C, greenhouse, Potsdam-Golm, Max Planck Institute for Molecular Plant Phyfresh sample of developing tuber was lyophilized before measurement of dry weight andecreased water content but improved tuber weight and starch content. Starch content wby different letters (a–c) are significantly different (P < 0.05), n = 4–6. DW, dry weight (g

greenhouse from April 2010 to August 2010 under LD conditions.The second and the third harvest plants were grown in green-houses in Hong Kong (Fig. S4) and in Golm (Table 2 and Fig. S5),respectively. All three high expression OE lines displayed earlierflowering than the WT (Table 1). Floral and tuberization transitionsin potatoes are controlled by two different FT-like paralogues thatrespond to environmental cues [17]. As OE lines exhibited earlierflowering (Table 1) and produced more tubers (Table 2), it wouldbe interesting to examine if FT expression is driven by high sugarsupply from source tissues.

Genetic manipulation of sucrose transporter (SoSUT1) in potatohas been documented to result in a shift in carbon partitioning inboth leaves and tubers and improved assimilation rates [18]. OurOE lines generally exhibited higher photosynthetic rates (10–20%) than the WT (Fig. 2) and higher StSUT1 expression level(Fig. 4). After 4 months of growth in the greenhouse, tubers werecollected when the plants were totally senescent. In both harvests(Figs. S4 and S5), all three high AtPAP2 expression lines (OE1, 4,and 7) produced more tubers than the WT and the biomass ofaboveground organs was also higher in the OE lines (Table 2). Ina separate experiment using a different potato cultivar (S. tubero-sum cv. Desirée), highly similar results were obtained, suggestingthat the observation was independent of the cultivar. Comparedwith the WT, overexpression of AtPAP2 increased the tuber yield

E1 OE4 OE7 OE10

3 ± 0.8d 44 ± 0.9d 45 ± 1.3d 73 ± 0.8c

E2 OE3

7.7 ± 1.4b 38 ± 1.2b

o December 2010 in greenhouse (SD, Hong Kong). Potatoes (S. tuberosum cv. Desirée)arkness, 20 �C, greenhouse, Potsdam-Golm, Max Planck Institute for Molecular Plantlues marked by different letters (a–c) are significantly different (P < 0.05), n = 4–6.

OE1 OE4 OE7 OE10

3 ± 0.8b 3b 2.8 ± 0.5ab 2.0 ± 0.8ab

18.6 ± 2.9cd 15.3 ± 3.0bc 21.6 ± 2.8d 11.4 ± 5.2ab

55.9 ± 16.4c 45.9 ± 9.1bc 60.2 ± 16.7c 20.7 ± 8.7a

1.4 ± 0.4b 0.5 ± 0.2a 1.1 ± 0.3b 0.4 ± 0.3a

5.1 ± 0.6c 3.7 ± 0.7b 4.6 ± 0.6bc 2.9 ± 0.5a

2.7 ± 0.2ab 2.3 ± 0.7ab 2.7 ± 0.7ab 1.9 ± 0.3a

2.4 ± 0.5c 1.6 ± 0.4ab 2.0 ± 0.2bc 1.1 ± 0.3a

OE3 OE4 OE7 OE10

15 22 18 20270.2 300.9 275.9 241.3

OE1 OE2 OE3

11.5 ± 1.8b 9.8 ± 0.8b 10.3 ± 1.4b

281.4 ± 17.8b 267.8 ± 14.5b 252.5 ± 17b

53.9 ± 3.4b 52.7 ± 2.9b 48.0 ± 3.2b

806.6 ± 4.1b 803.0 ± 13.2b 811.1 ± 15.2b

416.8 ± 9.9b 429.9 ± 8.9b 419.13 ± 5.3b

the decay of the shoots following a 4-month growth period. The second generationthird generation was grown in long day condition in University of Potsdam, Golm.ber 2013 to February. 2014 [16 h light period (100–200 lE m�2 s�1), 25 �C and 8 hsiology]. Throughout the light–dark cycle relative humidity was kept at 50%. 100 mgd water content. Compared with that of the wild type, tubers of OE lines showedas measured by 10 mg fresh potato tubes. Within each column, the values marked

); FW, fresh weight (g).

Fig. 2. Assimilation rates of potato plants. Assimilation rates (lmol CO2 m�2 s�1)were measured on leaves 3–4 from the top of 65- to 68-day-old plants. Assimilationrates at different actinic light intensities (0–1000 lmol photons m�2 s�1) werenormalized to leaf area.

Y. Zhang et al. / FEBS Letters 588 (2014) 3726–3731 3729

per plant �2 to 3-fold in S. tuberosum var. Bintje and 2-fold in S.tuberosum cv. Desirée, respectively. The increase in tuber yieldwas due to an increase in dry weight and starch content (Table 2).In total, the five greenhouse trials all produced highly similarresults.

Increased sink demand (via systemic signals) and decreasedphotoassimilate levels in source leaves (via an alleviated feedbackrepression of photosynthesis by sugar sensing) can both enhancephotosynthetic activity [19]. Therefore, a high performance liquidchromatography method was next used to evaluate sugar contentsof leaves immediately following the measurement of the rate ofphotosynthesis. The sucrose contents of all OE lines weredecreased by �60%, with glucose and fructose also being decreased(Fig. 3). These results suggested an inverse relationship betweenAtPAP2 expression and the contents of sucrose, glucose andfructose in potato leaves. For comparison we analyzed the sugar

Fig. 3. Sugar contents in leaves, petioles and tubers of second generation plants.The leaves of OE lines exhibited low sucrose, glucose and fructose contents.However, the petioles of OE lines contained higher sucrose, glucose and fructosecontent and the tuber OE lines shows higher sucrose, glucose and fructose contents.Within each column, the values are the fold change of OE lines compare with wildtype. DW – dry weight, FW – fresh weight.

content of the petioles (phloem) of the top 2 to 3 fully-expandedleaves of 68-day-old potatoes. In contrast to the leaves, the sucroselevels of the petioles were increased by 1.5 to 2-fold in the petiole,while glucose and fructose contents increased moderately (Fig. 3).The higher concentration of sucrose in the petioles is consistentwith its higher sucrose transport activity. Overexpression ofSoSUT1 in potato resulted in a shift in carbon partitioning in bothleaves and tubers and improved photosynthesis rate [18].Improved rates of sugar efflux via the leaf petioles would stimulatepetioles loading and lower mesophyll carbohydrate levels and thusrelieve inhibition of photosynthetic activity. Consistent with thishypothesis the sugar levels in tubers of the three higher AtPAP2overexpression lines tubers were greatly increased (Table 2).

The rate of sucrose synthesis controlled by SPS was shown tocorrelate with the rate of photosynthesis and with the rate ofexport from leaves [20]. The activity of the SPS is inhibited by bind-ing with the 14-3-3 protein which is regulated by the SnRK1 pro-tein kinase. In previous studies transgenic Camelina andArabidopsis overexpressing AtPAP2 exhibited higher SPS activityin the leaves [5]. Moreover, transgenic tomato, potato, Arabidopsisand tobacco expressing various SPS genes were documented toexhibit increased biomass and photosynthesis rate. As shown inTable 3, SPS activity was enhanced in the leaves of AtPAP2 OEplants in both optimal Vmax and limiting Vlimit capacities. Thus,both a higher photosynthetic rate (Fig. 2) and an increased SPSactivity in OE lines can provide more sucrose for growth. Westernblotting analysis using an anti-SPS antibody (Agrisera, Sweden)indicated that SPS accumulation was remarkably enhanced in theOE lines. However, the protein expression levels of nitrate

Fig. 4. Western blotting analysis of proteins involved in the sucrose synthesis. Totalsoluble protein extracts were from 70-day-old second generation plants in themiddle of day. Vector is the transgenic potato lines by empty vector pBA002; SPS,sucrose phosphate synthase; NR, nitrate reductase; StSUT1, Solanum tuberosumsucrose transporter 1; cFBPase, cytosolic fructose-1,6-bisphosphatase; Anti-cFB-Pase, anti-NR and anti-SPS antibodies were obtained from Agrisera; Anti-StSUT1antibody was provided by Dr. Christina Kuhn (Humboldt University Berlin,Germany); AtPAP2, anti-AtPAP2 specific antibody; Anti-14-3-3 antibody was fromProf. Carol MacKintosh. Excess 14-3-3 recombinant protein was loaded on the GSTbeads and mixed with 0.5 mg total plant protein extracts, the washed and boundproteins were eluted for Western blotting analysis using the anti-SPS antibody.

Table 3SPS Enzyme activity analysis.

lM sucrose/lg protein/h WT Vector OE1 OE4 OE7 OE10

Vmax 139 ± 4a 147 ± 4ab 168 ± 6c 163 ± 13c 162 ± 9c 155 ± 6c

Vlimit 73 ± 3a 83 ± 5b 97 ± 2c 93 ± 7c 93 ± 6bc 90 ± 5bc

The third and the fourth fully mature leaves were used for SPS enzyme activity analysis after the photosynthesis measurement. Within each column, the values marked bydifferent letters (a, b) are significantly different (P < 0.05), n = 5.

Table 4Relative transcriptional levels of sucrose transporters.

WT Vector OE1 OE4 OE7 OE10

StSUT1 1.00 ± 0.01a 1.31 ± 0.01b 1.53 ± 0.01b 1.07 ± 0.02a 1.21 ± 0.01ab 1.03 ± 0.01a

StSUT2 1.00 ± 0.01a 1.02 ± 0.01a 0.36 ± 0.01b 0.25 ± 0.01b 0.23 ± 0.01b 0.07 ± 0.01c

StSUT4 1.00 ± 0.02a 0.74 ± 0.01ab 0.57 ± 0.02b 0.58 ± 0.01b 0.45 ± 0.02b 0.08 ± 0.01c

The third and the fourth 70-days-old mature leaves were used for real-time PCR analysis. RQ, the gene relative expression to WT was calculated using the equation: 2�(44CT).Within each column, the values marked by different letters (a–c) are significantly different (P < 0.05), n = 3.

3730 Y. Zhang et al. / FEBS Letters 588 (2014) 3726–3731

reductase (NR) and fructose bisphosphatase (FBPase) wereindistinguishable from the WT (Fig. 4). Moreover, the levels of14-3-3 protein and the amount of phosphorylated SPS that wascapable of binding 14-3-3 were unaltered in the OE lines. Theseresults indicated that only unphosphorylated SPS was greatlyenhanced in the OE lines, which would be anticipated to result ina far greater in vivo SPS enzyme activity. To determine if therewas direct interaction between AtPAP2 and SPS, the crude proteinextracts of potato leaves were immuno-precipitated by the anti-SPS antibody. The bound proteins were then detected by an anti-AtPAP2 antibody via Western blotting. No direct interactionbetween AtPAP2 and SPS was observed (Fig. S6). Besides, theexpression level of SnRK1 was not significantly changed (Fig. 4),and no protein interaction between SnRK1 and AtPAP2 could bedetected by the yeast two-hybrid assay (data not shown). Hence,the higher SPS activity in the OE lines was attributed to a higherexpression level of SPS protein, rather than through activation bypost-translational modification. Therefore, overexpression ofAtPAP2 appears to indirectly regulate the expression and enzymeactivity of SPS, thus affecting sucrose synthesis, flower time andtuberization.

To examine how AtPAP2 overexpression might affect sugarpartitioning in leaves and petioles, the expression levels of sugartransporters in leaves were examined. Although the expressionlevel of StSUT1 transcript did not change significantly, its proteinlevel was significantly elevated in the OE lines. StSUT1 is essentialfor long-distance transport of sucrose and plays a role in phloemloading in mature leaves [21]. Its higher expression in the leaf ofOE lines may lead to a higher sucrose level in petioles and tubersbut a lower leaf sucrose content (Fig. 3). In contrast, the transcrip-tion levels of StSUT2 and StSUT4 were greatly decreased (Table 4).Sucrose transporters are known to be regulated by phosphoryla-tion [22], however, co-immunoprecipitation assays did not revealany direct interaction between AtPAP2 and StSUT1 (Fig. S6).

Many plant PAPs mediate phosphorus acquisition and redistri-bution based on their ability to hydrolyze phosphorus compounds[3]. AtPAP2 is a phosphatase anchored on the outer membrane ofchloroplasts and mitochondria [4]. Theoretically, overexpressionof a phosphatase in cytosol may supply additional phosphate as acounter-exchange substrate for the triose phosphate/phosphatetranslocator (TPT) on chloroplasts to facilitate higher export of tri-ose phosphates to cytosol for sucrose synthesis. This, however, isunlikely the reason for higher tuber yield in the AtPAP2 OE lines.First, the Pi content was not significant changed in our OE lines(Table. S2). Second, overexpression of a soluble Escherichia colipyrophosphatase, (PPase) which could generate more phosphates

by hydrolyzing pyrophosphates, in the cytosol of leaf cells didnot lead to increase in tuber yield [23].

Changes in carbon flow and sink/source activities could affectfloral, architectural, and reproductive traits of plants. In potato,the tuber yield could be improved by simultaneously geneticallymodifying source and sink strengths by sucrose transporter [18].The potato sink strength is defined as the ability to attract photo-assimilates, and the sink strength of growing potato tubers wasassumed to be limited by metabolism and/or starch synthesis[24]. Sink strength had been concluded to be a more important fac-tor than source strength on tuber yield [23]. In that study, the tuberyield was enhanced in the PGN and AGN lines. Sugar contents inphloem were higher but the leaf starch was lower in the transgeniclines, reflecting an increase in sink strength in tuber could redis-tribute the carbohydrates from source (leaf starch) to tuber starch.The authors also produced transgenic lines with enhanced sourcestrength by overexpressing E. coli PPase in potato leaf [23]. Whilesome transgenic lines exhibited higher PPase activities, higher sug-ars and lower starch in leaves, the tuber yield did not increase. Theauthor therefore concluded that tuber yield is sink-limited, andthat an additional enhancement of source capacity could furtherincrease yield [23]. However, it should be noted that the photosyn-thesis rates of the AGN lines were not enhanced, and the sugar con-tents in the petioles and tubers of the PPase overexpression lineswere not measured. Thus the potential of increasing tuber yieldby enhancing photosynthesis rate in source leaf and/or sugarexport from source leaf cannot be ruled out.

In this study, overexpression of AtPAP2 could improve sourcecapacity by improved photosynthesis rate, elevated SPS activity,and sugar efflux rate in the leaves. Increased photosynthesis inthe source tissues could also potentially improve the adenylatepools in the potato leaves [25]. Increased adenylate pools have pre-viously been demonstrated to increase the potato tuber yield inexperiments in which the activity of the plastidial adenylate kinasewas altered [26]. ATP and adenylate pools were shown to be signif-icantly increased in the rosette leaves of AtPAP2 OE Arabidopsislines [9]. Meanwhile, the expression levels and enzyme activityof SPS were also significantly elevated in the AtPAP2 OE lines,which could be expected to supply more sucrose for export andanabolism. Similar to the situation observed following up-regulated expression of the sucrose transporter [18], overexpres-sion of AtPAP2 resulted in an increased sugar efflux, improved pho-tosynthesis and faster plant growth rate. In addition, coupled to theaccumulation of soluble sugars in the tuber, the total tuber yieldand tuber starch content were improved in our OE lines. These dataindicate an enhanced sink strength. We believe both (limited by

Y. Zhang et al. / FEBS Letters 588 (2014) 3726–3731 3731

sink and promoted by source) are valid as nature in generally doesnot follow all-or-none principle. We deduced that overexpressionof AtPAP2 improves the source capacity and sink strength of pota-toes by indirectly regulating the expression of SPS and sucrosetransporters.

Funding

This project was supported by the General Research Fund(HKU772710M) of the HKSAR, China.

Acknowledgements

We would like to thank Prof. M.L. Chye (HKU) for providingpotato. We are thankful to Prof. Steven F. Chen (HKU) and Dr. CliveS.C. Lo (HKU) for providing the HPLC machine. Youjun Zhangwishes to thank HKU for the postgraduate studentship and DAADscholarship.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.febslet.2014.08.019.

References

[1] Smeekens, S., Ma, J., Hanson, J. and Rolland, F. (2010) Sugar signals andmolecular networks controlling plant growth. Curr. Opin. Plant Biol. 13, 273–278.

[2] Wind, J., Smeekens, S. and Hanson, J. (2010) Sucrose: metabolite and signalingmolecule. Phytochemistry 71, 1610–1614.

[3] Kuang, R., Chan, K.-H., Yeung, E. and Lim, B.L. (2009) Molecular andbiochemical characterization of AtPAP15, a purple acid phosphatase withphytase activity, in Arabidopsis. Plant Physiol. 151, 199–209.

[4] Sun, F., Carrie, C., Law, S., Murcha, M.W., Zhang, R., Law, Y.S., Suen, P.K.,Whelan, J. and Lim, B.L. (2012) AtPAP2 is a tail-anchored protein in the outermembrane of chloroplasts and mitochondria. Plant Signal. Behav. 7, 927–932.

[5] Zhang, Y., Yu, L., Yung, K.-F., Leung, D.Y., Sun, F. and Lim, B.L. (2012) Over-expression of AtPAP2 in Camelina sativa leads to faster plant growth andhigher seed yield. Biotechnol. Biofuels 5, 1–10.

[6] Sun, F., Suen, P.K., Zhang, Y., Liang, C., Carrie, C., Whelan, J., Ward, J.L., Hawkins,N.D., Jiang, L. and Lim, B.L. (2012) A dual-targeted purple acid phosphatase inArabidopsis thaliana moderates carbon metabolism and its overexpressionleads to faster plant growth and higher seed yield. New Phytol. 194, 206–219.

[7] Murashige, T. and Skoog, F. (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15, 473–497.

[8] Chincinska, I., Liesche, J., Krugel, U., Michalska, J., Geigenberger, P., Grimm, B.and Kuhn, C. (2008) Sucrose transporter StSUT4 from potato affects flowering,tuberization, and shade avoidance response. Plant Physiol. 146, 515–528.

[9] Sun, F., Liang, C., Whelan, J., Yang, J., Zhang, P. and Lim, B.L. (2013) Globaltranscriptome analysis of AtPAP2-overexpressing Arabidopsis thaliana withelevated ATP. BMC Genomics 14, 752.

[10] Beaujean, A., Sangwan, R., Lecardonnel, A. and Sangwan-Norreel, B. (1998)Agrobacterium-mediated transformation of three economically importantpotato cultivars using sliced internodal explants: an efficient protocol oftransformation. J. Exp. Bot. 49, 1589–1595.

[11] Sambrook, J., Fritsch, E. and Maniatis, T. (1989) Molecular Cloning: ALaboratory Manual, Cold Spring Harbor Laboratory Press, New York. pp.931–957.

[12] Baroja-Fernandez, E., Muñoz, F.J., Saikusa, T., Rodriguez-Lopez, M., Akazawa, T.and Pozueta-Romero, J. (2003) Sucrose synthase catalyzes the de novoproduction of ADPglucose linked to starch biosynthesis in heterotrophictissues of plants. Plant Cell Physiol. 44, 500–509.

[13] Fernie, A.R., Roscher, A., Ratcliffe, R.G. and Kruger, N.J. (2001) Fructose 2, 6-bisphosphate activates pyrophosphate: fructose-6-phosphate 1-phosphotransferase and increases triose phosphate to hexose phosphatecycling in heterotrophic cells. Planta 212, 250–263.

[14] Baxter, C.J., Foyer, C.H., Turner, J., Rolfe, S.A. and Quick, W.P. (2003) Elevatedsucrose-phosphate synthase activity in transgenic tobacco sustainsphotosynthesis in older leaves and alters development. J. Exp. Bot. 54,1813–1820.

[15] Emi, T., Kinoshita, T. and Shimazaki, K. (2001) Specific binding of vf14-3-3aisoform to the plasma membrane H+-ATPase in response to blue light andfusicoccin in guard cells of broad bean. Plant Physiol. 125, 1115–1125.

[16] Moorhead, G., Douglas, P., Cotelle, V., Harthill, J., Morrice, N., Meek, S., Deiting,U., Stitt, M., Scarabel, M. and Aitken, A. (1999) Phosphorylation-dependentinteractions between enzymes of plant metabolism and 14-3-3 proteins. PlantJ. 18, 1–12.

[17] Navarro, C., Abelenda, J.A., Cruz-Oró, E., Cuéllar, C.A., Tamaki, S., Silva, J.,Shimamoto, K. and Prat, S. (2011) Control of flowering and storage organformation in potato by FLOWERING LOCUS T. Nature 478, 119–122.

[18] Leggewie, G., Kolbe, A., Lemoine, R., Roessner, U., Lytovchenko, A., Zuther, E.,Kehr, J., Frommer, W.B., Riesmeier, J.W. and Willmitzer, L. (2003)Overexpression of the sucrose transporter SoSUT1 in potato results inalterations in leaf carbon partitioning and in tuber metabolism but has littleimpact on tuber morphology. Planta 217, 158–167.

[19] Ainsworth, E.A. and Bush, D.R. (2011) Carbohydrate export from the leaf: ahighly regulated process and target to enhance photosynthesis andproductivity. Plant Physiol. 155, 64–69.

[20] Huber, S.C. and Huber, J.L. (1992) Role of sucrose-phosphate synthase insucrose metabolism in leaves. Plant Physiol. 99, 1275–1278.

[21] Kühn, C., Hajirezaei, M.-R., Fernie, A.R., Roessner-Tunali, U., Czechowski, T.,Hirner, B. and Frommer, W.B. (2003) The sucrose transporter StSUT1 localizesto sieve elements in potato tuber phloem and influences tuber physiology anddevelopment. Plant Physiol. 131, 102–113.

[22] Krügel, U., He, H.-X., Gier, K., Reins, J., Chincinska, I., Grimm, B., Schulze, W.X.and Kühn, C. (2012) The potato sucrose transporter StSUT1 interacts with aDRM-associated protein disulfide isomerase. Mol. Plant 5, 43–62.

[23] Jonik, C., Sonnewald, U., Hajirezaei, M.-R., Flügge, U.-I. and Ludewig, F. (2012)Simultaneous boosting of source and sink capacities doubles tuber starch yieldof potato plants. Plant Biotechnol. J. 10, 1088–1098.

[24] Zrenner, R., Salanoubat, M., Willmitzer, L. and Sonnewald, U. (1995) Evidenceof the crucial role of sucrose synthase for sink strength using transgenic potatoplants (Solanum tuberosum L.). Plant J. 7, 97–107.

[25] Chakraborty, S., Chakraborty, N., Agrawal, L., Ghosh, S., Narula, K., Shekhar, S.,Naik, P.S., Pande, P., Chakrborti, S.K. and Datta, A. (2010) Next-generationprotein-rich potato expressing the seed protein gene Am A1 is a result ofproteome rebalancing in transgenic tuber. Proc. Natl. Acad. Sci. 107, 17533–17538.

[26] Regierer, B., Fernie, A.R., Springer, F., Perez-Melis, A., Leisse, A., Koehl, K.,Willmitzer, L., Geigenberger, P. and Kossmann, J. (2002) Starch content andyield increase as a result of altering adenylate pools in transgenic plants. Nat.Biotechnol. 20, 1256–1260.